The underlying basis of systemic sclerosis, scleroderma, is unknown. Cultured dermal fibroblasts from scleroderma patients overexpress extracellular matrix components, thus retaining a feature of scleroderma skin in the culture model. We used differential display and hybridization to large arrays of expressed sequence tags to compare gene expression in scleroderma and healthy fibroblasts. Our recently published data show that protease nexin 1, a protein that regulates matrix metabolism, is expressed in scleroderma skin but not in skin from healthy individuals. Because protease nexin 1 is known to inhibit the activation of collagenase, and because we have shown that protease nexin 1 induces collagen transcription, we created transgenic mice containing the human protease nexin 1 cDNA in a cytomegalovirus transcription unit. Part of the current proposal is to examine these mice as a potential model of fibrotic disease. We recently found another gene with more dramatic differential expression. Hybridization to large arrays of expressed sequence tags demonstrated that heat shock protein 90 (hsp90) is overexpressed in scleroderma fibroblasts. Northern analysis showed that hsp90 is highly expressed in scleroderma fibroblasts and not detected in healthy fibroblasts. Overexpression in healthy cells or heat shock itself caused a significant increase in endogenous collagen message. Overexpression of hsp90 also causes a 3.6-fold reduction in collagenase promoter activity (MMP1). Furthermore, a specific inhibitor of hsp90, geldanamycin, obliterates TGFbeta-induced collagen transcription. Hsp90 is known as a molecular chaperone. The chaperone activity of hsp90 is essential to the normal function of the hormone receptor. TGFbeta activates a receptor system, which in turn causes phosphorylation of a cytoplasmic protein called Smad. Phosphorylation of Smad causes its transport to the nucleus where it binds to a specific transcriptional regulatory sequence. Our recent novel finding is that hsp90 is a component of the Smad signaling complex. The second aim of this proposal is therefore to more firmly understand the overexpression of hsp90 in scleroderma skin. The final aim in this proposal is to define the nature of the interactions between Smad, hsp90 and related proteins in the Smad signaling complex and thus to understand how hsp90 functions in regarding TGF-beta signaling.